`Error fetching route from Graphhopper Error: Network response was not ok`

I tried a couple times and got this in the U.S. In Europe I picked a couple random points and it worked fine.

Unlikely. Even with turn costs enabled 256GB (or less) are sufficient. You could also try to disable CH as for bike often no long routes are required (you could disable them). Here we have written down a few more details: https://www.graphhopper.com/blog/2022/06/27/host-your-own-wo...

I thought I would be able to compute the graph with 64GB of ram but it kept crushing before CH and LM stage. After switching to a 128GB instance, it finally worked, hitting around 90GB at peak memory usage. For context, I was using 3 profiles - one with CH and two with LM, plus elevation data and used all of the tips from deploy.md

Maybe you already considered, but there are a number of collection libraries out there that are optimized for holding Java primitives and/or for very large sets of data, which could help you save significant memory. Eclipse Collections [0] and Fastutil [1] come to mind first, but there are many out there [2]

[0] https://github.com/eclipse-collections/eclipse-collections [1] https://fastutil.di.unimi.it/ [2] https://github.com/carrotsearch/hppc/blob/master/ALTERNATIVE...

Yes, definitely.

> I thought I would be able to compute the graph with 64GB of ram but it kept crushing before CH and LM stage.

For normal GraphHopper and just the EU the 64GB should be more than sufficient.

Cool idea, i'd love to try it but honestly i'd love it on my phone for the aforementioned reasons!

---

Edit: Some follow up remarks (From Chrome on a Mac)

1. I found it a bit confusing when creating a route. I assumed it would let me do address completion. The UI seems to just be naming my route instead, i think if you're going to borrow visually from Google maps you should follow their patterns.

2. I was able to add my first point (my home), but adding a second point never worked.

3. The click mechanic is a bit odd. I expect clicking elsewhere when the menu is open (add point/close) that the menu would go away - rather than re-appear in my new click location.

4. The map centred me over Europe, i'm in Canada however. Some GeoIP lookup here could give a better experience.

5. Re:point 2, this seems to work when i selected random locations in Europe. I assume this is a data issue? So the real issue here (apart from the routing not working) is there's no feedback on the UI when the routing fails.

6. Refreshing my screen loses my route. Any chance you could save it to local storage or something? Would be amazing if i could create this route on my desktop and then send a link to my phone (once mobile is supported?)

Regarding state: it's only a matter of changing the blacklist array in redux persist config. Right now I have all global state blacklisted because I noticed a small bug with rendering the map layers when the state is persisted and I didn't have time to fix it but in the next version it will be persisted in localstorage.

I couldn't imagine trying to build a complex route on a tiny screen with a crappy phone UI. I am starting to feel like an old man yelling at clouds here. Why in the hell wouldn't you prefer to use a nice monitor with a mouse for a complex task like this?

I nearly always plan routes on my phone. After all, it's the device I use for navigation attached to the stem of my bike.

> Why in the hell wouldn't you prefer to use a nice monitor with a mouse for a complex task like this?

The task isn't very complex with a decent UI. See for example the open-source app OSMAnd (Android & iPhone). Also, I am rarely at my desk, but I always have my phone on me. There can also be the issue of "Neat, I made a route on a big screen; now, how do I get it on my phone for real-time navigation?" depending on the app you are using.

Thia isn't to say you're wrong. I much prefer my monitor and mouse for anything of reasonable complexity. But I increasingly find myself in the minority in that regard among people I know. Hell, I even find myself in the minority for using a laptop for tasks sometimes.

There are simply some things where a phone screen is just too small to use efficiently, and the fingers sometimes aren't just a good, precise enough input device.

Though I am a bit irritated by the brightness of the route on your site, it lacks contrast with respect to the surrounding map.

BTW, how is that routing done? Like which is the used routing engine and is it done server-side or in the browser?

---

It does seem to have a bug, where, when deleting a waypoint, the mouse still is in "create waypoint mode", and any mouse-down on the map, including for panning, results in adding a waypoint.

It really reminds me of https://trailrouter.com/ which I've found amazing for finding new running routes.

It would be amazing to have similar functionality to suggest cycling routes of a certain distance that are mostly on bike trails i.e. suggest to me a 40km circular/out-and-back bike route from my house.

One of the most frustrating things for me with snapping to official paths is not being able to modify it for common workarounds (for instance, going across the Golden Gate bridge, everyone takes a shortcut through a parking lot, but every map routing platform I have used forces me to go the official route and messes up my nav)

I'm around hill country TX now where there is typically a patchwork of bike lanes segments that start and stop without much attention given to continuity of (sub)urban planning, walkability, safety, or design consideration for non-motorized users.

I too am interested in linking up good bike trails. Mostly for the east bay gravel systems. Today, I save GPX or geojson from routes I find on Strava and import into a map client (CalTopo). It’s a okay solution but my problem is in finding more alternative routes.

I added a similar feature to cycle.travel (my site) a few months back. 5% of the work was writing the code to route only on waymarked routes, 95% was writing code to "heal" the accidental breaks in OSM bike route data...

Doing OSM editing from Mapillary imagery in various US states, I find that the roads that have roadside signage as “official bike routes”, predate modern safe(r) bike infrastructure. These are often highways that frankly look like death traps: no shoulder, high speed limits. Especially when so many North American long-haul cyclists now are riding “bikepacking” setups, it may be preferable for a router to prioritize quiet agricultural tracks and other unpaved terrain over the supposedly official bike route.

cycle.travel treats routes in the States differently to those in Europe for precisely this reason, and there are some routes which get absolutely zero uplift because many of them are so bad (looking at you, East Coast Greenway).

https://brouter.de/essbee/

As for Graphhopper, I ran into some challenges during the Europe-wide import stage. It turns out 64GB of RAM wasn’t enough, so I ended up spinning up a 128GB instance on AWS. After tweaking some config settings and following the deployment guide, I finally got it working. I also had to change the source code a bit to link each "official bike route" edge to its corresponding OSM relation info but I managed to get it working in the end(using KVStorage and KValues).

For now, my planner is pretty basic, so I don’t want to bombard you with beginner-level questions. But if something more complex comes up down the road, I’ll definitely take you up on your offer to reach out. Thanks again—I really appreciate your help and generosity!

We use graphhopper for everything, with customized profiles. We add data to the weekly planet PBF files from the OSM project for additional routing data not present in OSM.

This morning, I picked up an affordable m920q with 64GB of RAM and I’m planning to use it for hosting GH and maybe vector tiles too. If I run into any issues I can’t figure out quickly I’ll reach out but I don’t want to bug you with basic questions.

I was looking at Rails (API mode + partial asset pipeline) + Next.js for a project to start, and might entertain elixir or rust when scalability of slow areas became a problem. Sometimes I think there will be eventually a system of LLMs that will figure out how to manage and configure operating systems, write mobile native and web apps, and administer databases. And then they'll eventually rewrite programming languages and design silicon to be better for themselves, so that they will only need a part time human to clean and maintain the robot security guard.

Sometimes it's actually better to take the road, or you end up on a windy route that's near impossible to follow, debris everywhere, no right of way, etc.

I would like to try this with a "prefer offroad" option.

Mobile needs some work. Are you planning on open sourcing this? I'm a mobile dev that might be able to clean that up a little. Good candidate for React Native.

It would be great™ if it could drag new mandatory points and choose alternative routes. Also, switching from administrative to with satellite view would be helpful.

One thing that would be very useful is to color the segments on the map based on the waytype. The proportions are given in the summary but unless I already know the route I can't tell where exactly those segments of difficult cobblestone are.

[0] - https://cyclers.app/

[1] - https://www.cyclestreets.net/

So far, every planner I have tried from this thread has just been frustrating in one way or another. It seems very hard for most to understand things like road+MTB in one ride or that the map shows that the road doesn't go through, but I can simply walk through a gate that a car can't.

This looks interesting and the mobile app seems well made, I like that it suggests multiple routes. Currently I just use Google Maps and it sucks here as well.

Had to customize https://brouter.de/brouter-web to make something workable. But i's annoying to paste script, route, export gps, send to phone, import in komot just for a 30min ride. (Can share if anyone is interesed. Just be very respectful please, it does not avoid footpaths either)

https://rc.nau.edu/cranc/?profile=ibc&layer=OpenStreetMap

The only feature I'd want is to be able to print the route on paper maps. I wonder if print CSS could be used to create pages of the whole route.

I have personally had good luck with my Garmin watch. If I start out near a major trail, it tends to suggest routes along that trail. I’m able to input a desired length such as 10 miles and then it suggests multiple routes that are close to that length and use major trails. My biggest complaint with the Garmin is that it tends to suggest the same routes every day.

Would be cool to also show the elevation for a given segment along with its distance (in the route trail list).

And maybe when you click on a segment in the map, it should highlight that segment in the trial list so you see the length of that segment.

I realize you're just using existing map data, but some bike routes seem to be broken up with very small (couple km) gaps (shown in grey) -- is that really the case? (see Loire a Velo 4 for example)

Constructive feedback: at least on safari, it was not as smooth to navigate as other map apps, and also a bit cumbersome to add waypoints, etc.

A small but important thing - there is a typo in the “activity” spelling.

Hope that helps, and best of luck with it!

Boulder, Colorado for reference.

Not sure about the UK but I've cycled quite a lot around here in Europe (12 countries so far) and from my experience, if you don't follow the official bike trails you're missing a lot. They are usually nice scenic and quiet roads with good (asphalt) surfaces. I'm talking about trails like the Alpe Adria Cycle path. There are tons of them in mainland Europe.

But you're right - if the surface is bad, then it doesn't make sense to follow the trail. I will add a feature that will highlight the selected surface on the line (similar to what Komoot already does) so you can better investigate the route and make sure the surface is alright.

I'm also planning to add a feature where users could rate the track so you could filter out the bad tracks based on other people's reviews.

https://brouter.de/essbee/

Select it in in upper left menu - then fine tune the script on the right.

UI is not as good but, there is very useful tools like camping, train station, the ability to split your trip in steps ect.

Edit: I see now that the cycle.travel developer is here in the thread :)

BTW Typo: 'Acitvity type' in the UI should be 'Activity type'.

I've been bikepaking for two years now and I wish I had this tool

I've used komoot in the past but I was never satisfied with the paths it suggested and had to rely on local guides

I've checked with some of the past trips I id and it guesses correctly all of the best paths

        POST https://routes.trailimap.com/route/ 400 (Bad Request)

        Error fetching route from Graphhopper Error: Network response was not ok

        Uncaught (in promise) TypeError: Cannot read properties of undefined (reading 'paths')
    at g (page-c19a2a89477d3054.js:1:35496)

May be useful to you or others in this thread. We lean pretty heavily into the "cycling power user" market segment, the feature set isn't always the most discoverable but it's quite comprehensive if you put in the up-front time to learn the tooling (similar to a lot of other specialty mapping apps out there - caltopo, fatmap (rip) etc)

I'd spend up to a $100 once on a killer tool that meets all my needs in this space.

This subscription first crap is way out of hand.

of course; this is primarily a US audience (maybe parts of Europe)

* excellent, almost entirely bug-free routing on mobile

* heatmap data, because maps aren't entirely up-to-date

* multiple map styles, so you can pick what works best for your workflow and the country you're in

* easy GPX file export, I use it all the time with the bike computer (every day on tours)

* collection management, especially useful when I make per-day routes for a tour

* a healthy trial period so you can actually test it out and learn it

Basically it's just an excellent app (and site) that works reliably across every supported platform, that isn't full of spammy upselling garbage, that is clearly made by a competent team of developers who care deeply about the product they make.

Every tech product should be made like this. A lot of tech products used to be like this before enshittification really took off in the last 5-10 years.

I'm more than happy to support a great product like this, as a bicycle tourist and frequent router over unfrequented trails and dirt roads in the mountains around me. For road riders in cities, it's probably a whole lot less useful. But there are a lot of bicycle riding use cases outside of 'road riders in cities' :-)

This is me as well. Thanks much for your perspective, it was helpful in me making the decision to jump in.

I do like the idea of supporting products and services that are staving off enshitificaiton.

We have 14 machines in a rack at a datacenter in PDX, and have focused on low hosting costs since we have historically been bootstrapped and margin sensitive. Redundant switches, 2 firewall/load balancers, 4 compute machines, 3 database machines, 5 storage servers. Single upstream network provider, about to be two sometime next year. Rack space + power + redundant network is about $3500 a month. Machines have an average service life of about 5 years.

Database machines cost about $45k for a set of three.

Storing user data is non-trivial - GPS track files add up when you get close to a billion of them, photos are also very large. We use a self-hosted ceph object storage cluster of 5 machines, about $100k of hardware. it's cheaper than 20k a month in S3 bills.

Our rack all-in is probably about $250k of equipment. 5 year service life, probably $5k a month amortized out. So, doing things as cheap as possible (I buy nvme SSDs for storage cluster off ebay, and am about to buy a couple arista 100gbe switches from ebay as well) we are somewhere around $8500 a month on hosting.

We use both google maps as well as self-hosted OSM based map and routing services. About half our map usage goes to google, by user preference, and we pay them about $20,000 per month for that. Our self-hosted OSM map stuff require 1tb of ram, fast disks, a ton of CPU cores. We host 10 different planet scale routing profiles via graphhopper, which take 3 days to build every week with updated data. They also host a vector maps stack which is much more efficient, taking about 3 hours a week to build.

My last estimate of an AWS bill for all the above was $30k a month, assuming some discounts. We have grown since then and I'd napkin us to be > $40k a month at this point.

We strive to minimize any costs for third party platforms. We do use amplitude for analytics, that's > $30k a year at this point. We do use an external email service for easy marketing emails, but the majority of our millions of emails a month are sent from our own mail servers, using an in-house system we made a decade ago that still works well. We try to minimize vendor lockin and costs, where it makes sense.

Most expensive part of the entire company of course is salaries, with of course an eye to developer and related salaries. We run pretty bare bones where possible, with a flat management structure with minimal overhead. Our total staff size is 32, of which 6 are full-time end user support.

A bit of a ramble, sorry, but there's a large amount of overhead to run a system like ours. We don't just make a one-time use desktop application, we have to continually provide storage and compute for all users. If we stopped that, the entire service would fall apart. So yes, a subscription makes sense in a case like ours. You can't do what we do with a desktop app. Plus, the entire world has switched to mobile for this sort of consumer application, which is an entire rat race of it's own. You can't just release a single purchase app and expect it to be maintained, it's a massive effort to keep up with mobile development just to maintain features, much less build anything new.

This post, combined with the fact that your planner actually allows me to force a path has won me as a customer. I'll also note that setting your yearly at 75% of the monthly is wise considering very low winter time usage.

For me, churn or not is going to come down to whether I can read critical details on your mobile app without having to put reading glasses on. (which is a factor that not even a $2t company like Apple can address properly)

One comment: I hope you don't start emphasizing the "social media" aspects, like Strava has done, to their detriment, in my opinion. That's what prompted me to finally delete my Strava account (after uploading my thousands of activities to ridewithgps.)

Congratulations to your team on keeping costs low and running a successful business out of a single rack.

I have also worked with gps data and run my own map server (for a small country) so I know it costs money, time and effort. And thanks for the breakdown!

Why should it be a subscription instead of a pay-once app? Maps have to update as the real world updates, and probably they have other features that can't be on-device.

I'm paying for $1 and $5 subscriptions (ko-fi) for things I like and use. But $10 is getting too high for a thing even though I use it every day. Within five months I'm already past the value of my bike. I need that money to change parts to be able to continue biking.

What I ususally do is a one-month subscription to support. Then I turn it off for the rest of the year.

For me to sign up with a 3rd party service to pay a subscription: yes, $10/month is probably the smallest fee I'd bother with.

To add a subscription through my Apple devices, where I can manage my apps in a single pane of glass and start/stop subscriptions at will, I'm fine with paying $10/year, $2/month, whatever.

One little feature that's great is the quick link to view Google Streetview images, and the only thing I'd change is to have it let me do that anywhere on the map (not just on the route) so that I can look at alternatives without dragging the route there first.

It's a mostly a style for OSM with hillshading applied. The Friedrich-Alexander-Universität Erlangen-Nürnberg is hosting a web-based "preview" of it at https://opentopomap.org. https://brouter.de/brouter-web and https://bikerouter.de also have it as a selectable map layer, and it can also be added to OsmAnd.

Ideally, I'd also like to highlight roads with diverters, circles, chicanes, bumps, tables, etc.

I have not found any navigation app that preferentially chooses streets with traffic_calming features, such as speed humps. I recently suggested that in OSMAnd and got no traction [0].

[0] https://github.com/osmandapp/OsmAnd/issues/21320

While purpose-built traffic calming infra is super nice to have, I'm somewhat more interested in the much more prevalent (at least, outside Europe and Asia) and also muc less expensive roadways which are traffic-calm by default, owing to natural chicanes, unimproved areas, one-lane bridges, etc.

Bogota, for example, has a fairly elegant network of unimproved roads connecting the parks. If you take a ride with a local, you see the bike infra much more acutely than what is expressed on gmaps, or even theoretically by looking at `traffic_calming==true`. A much more granular data type is required.